A Brief Discussion on Digital Signal Processor

fighting

A Brief Discussion on Digital Signal Processor [Copy link]

A Brief Discussion on Digital Signal Processor

Liu Yinbi Lu Bin Wang Liang

Abstract : This paper summarizes and analyzes digital signal processors, covers the development of digital signal processors, and discusses the advantages of digital signal processors in combination with their software and hardware characteristics, and puts forward the problems and challenges faced by digital signal processors at this stage. Keywords : DSP Digital signal processing system software and hardware characteristics

1. Development History of DSP
The basis of informatization is digitization. One of the core technologies of digitization is digital signal processing. The task of digital signal processing needs to be completed by DSP devices to a large extent. DSP technology has become a cutting-edge technology that has attracted increasing attention and has been rapidly developed. DSP can stand for digital signal processor (Digital Signal Processor) or digital signal processing technology (Digital Signal Processing).


The latter is a theoretical technology that needs to be turned into a practical product through the former. The combination of the two becomes a means to solve a practical problem and implement a certain solution, namely, digital signal processing solution (DSPS). This article focuses on the first explanation of DSP - digital signal processor. DSP is a special processor that performs high-speed real-time processing after converting analog signals into digital signals. Its processing speed is 10 to 50 times faster than the fastest CPU. In the context of today's digital era, DSP has become a basic device in the fields of communications, computers, consumer electronics, etc. Industry insiders predict that DSP will be the fastest-growing electronic product in the future integrated circuits and will become a decisive factor in the replacement of electronic products.
Before the emergence of DSP, digital signal processing could only be completed by MPU (microprocessor). However, the low processing speed of MPU cannot meet the requirements of high-speed real-time. Therefore, in the 1970s, some people proposed the theoretical and algorithmic basis of DSP. However, DSP only remained in textbooks. Even the developed DSP system was composed of discrete components, and its application field was limited to military and aerospace departments.
With the development of large-scale integrated circuit technology, the world's first DSP chip was born in 1982. This DSP device is made of micron process NMOS technology. Although the power consumption and size are slightly larger, the computing speed is dozens of times faster than MPU. It has been widely used in speech synthesis and codecs. The advent of DSP chips marks a major step forward in the miniaturization of DSP application systems from large systems. With the advancement and development of CMOS technology, the second generation of DSP chips based on CMOS technology came into being. Its storage capacity and computing speed increased exponentially, becoming the basis of speech processing and image hardware processing technology. In the late 1980s, the third generation of DSP chips came out, and the computing speed was further improved. Its application scope gradually expanded to the fields of communications and computers.
DSP developed fastest in the 1990s, and the fourth and fifth generation DSP devices appeared one after another. The current DSP belongs to the fifth generation. Compared with the fourth generation, it has a higher system integration, integrating the DSP core and peripheral components on a single chip. This highly integrated DSP chip not only demonstrates its capabilities in the fields of communications and computers, but also gradually penetrates into people's daily consumption fields, and has a very promising future.
II. Characteristics and advantages of DSP
1. Hardware characteristics
(1) DSP belongs to the Modified Harvard architecture, that is, it has two internal buses: data bus and program bus. The program and data storage space are separated, each with an independent address bus and data bus. Instruction fetching and reading can be performed simultaneously, and the current speed has reached 9 billion floating point operations per second (9000 MFLOPS).
(2) Pipeline operation is adopted. The execution of each instruction is divided into several steps such as instruction fetching, decoding, data fetching, and execution, which are completed by multiple functional units in the chip. It is equivalent to executing multiple instructions in parallel, thereby greatly improving the operation speed.
(3) Independent hardware multiplier. The multiplication instruction is completed in a single cycle, optimizing the large number of repeated multiplications in algorithms such as convolution, digital filtering, FFT, correlation, and matrix operations.
(4) Special instructions such as circular addressing and bit-reversed greatly improve the addressing, sorting and calculation speed in FFT, convolution and other operations. The time for 1024-point FFT is less than 1μs.
(5) Independent DMA bus and controller. There are one or more independent DMA buses that work in parallel with the CPU's program and data buses. Without affecting the CPU's operation, the DMA speed has reached more than 800Mbyte/s.
(6) Multi-processor interface. It allows multiple processors to work in parallel or serially to improve processing speed.
(7) JTAG (Joint Test Action Group) standard test interface (IEEE 1149 standard interface). It is convenient for on-chip online simulation of DSP and debugging under multi-DSP conditions.
2. Software Features
(1) Immediate addressing: The operand is an immediate number and can be directly obtained from the instruction. Example: MOV A, @0x16; send the constant 0x16 to register A.
(2) Direct addressing: For example, TI's TMS320 series chips divide the data memory into 512 pages, with 128 words per page. Set a data page pointer DP (Data Pointer) to point to a data page with 9 bits, plus a 7-bit page offset address to form a 16-bit data address. This helps to speed up the addressing speed.
(3) Indirect addressing: ① 8 auxiliary registers, an auxiliary register pointer specifies an auxiliary register arithmetic unit to perform 16-bit unsigned number operations, determine a new address, and load it into one of the auxiliary registers. ② The contents of the 8 auxiliary registers are quite flexible, and can be loaded, added, and subtracted from immediate numbers; addresses can be loaded from the data memory; and some displacement addressing can also be performed. ③ Due to the use of reverse bit, bit reverse addressing can be achieved.
(4) Unique multiplication instructions: Example: MAC X0, Y0, AX: (R0) +, X0 Y: (R4) + N4, YO This instruction commands the DSP56300: multiply the numbers in registers X0 and Y0, add the result to Acc A, load the value in the memory address pointed to by register R0 into register X0, load the value in the Y memory address pointed to by register R4 into register Y0, add 1 to the value of R0, and add the value of register N4 to R4.
The advantages of DSP systems based on digital signal processing compared with traditional analog signal processing systems:
(1) Simple and convenient interface. Due to the simple electrical characteristics of digital signals, it is easy to implement hardware interfaces when different DSP systems are connected to each other. On the data stream interface, each system only needs to follow a specific standard protocol. (
2) High accuracy and good stability. Digital signal processing is only affected by binary errors and limited word length, and the processing process does not introduce other noise, so it has a high signal-to-noise ratio. In addition, the performance of analog systems is greatly affected by the performance of component parameters, while digital systems are basically unchanged, so digital systems are easier to test, debug and mass produce.
(3) Easy programming, easy to implement complex algorithms. In DSP systems, DSP chips provide a high-speed computing platform, and system functions rely on software programming. When combined with modern signal processing theory and computational mathematics, complex digital signal processing functions can be realized.
(4) Easy integration. Modern DSP chips integrate the DSP core and its peripheral circuits on a single chip. This structure facilitates the design of portable, highly integrated digital products.
In addition, modern DSP chips, as programmable very large integrated circuit (VLSI) devices, implement digital signal processing functions through downloadable software and firmware. In addition to the calculation and control functions of ordinary microprocessors, DSP chips have also made great improvements in processor structure, instruction system and instruction flow design for real-time digital signal processing with high data transmission rates and intensive numerical operations.
III. Problems and challenges faced by DSPs
The increasingly mature DSPs still have many areas that need to be improved, and they also face many challenges.
(1) How to arrange the data flow reasonably so that it can be executed smoothly without conflict between the execution units of the DSP is still an important problem faced by DSP developers. Due to the complexity of the design, high-level programming languages cannot be used when mapping the algorithm to the specific target hardware of the DSP. Assembly language must be used and the parallel execution mechanism of the device must be clearly understood. This programming design limited to assembly language is the bottleneck for improving software development efficiency.
(2) There are still problems in parallel structure. In order to achieve higher throughput, more data bits must be processed in a specific unit time. VLIW technology represents instruction-level parallelism. Superscalar and super-pipeline structures also attempt to get more instructions in one instruction cycle. Data-level parallelism is represented by wider data words, vectorization, and data flow structures. Since the width of the data word is larger, more data can be processed per instruction cycle, increasing the number of data bits that can be processed per clock cycle. Task-level or transaction-level parallelism is reflected in multi-tasking, multi-threading, and multi-processor designs. These structures are expected to improve data processing throughput, but the increased data and instruction width and the resulting increase in data processing throughput come at a certain cost. When code density and data width match the application, they can help, but when the data word width is different from the processor, they will cause a lot of trouble.
(3) The large amount of available on-chip cache is becoming more and more important to the total throughput of the system, because the standard memory bus and interface can no longer support the gigabyte data transfer rate of each MAC in the system. Whether the rest of the system can match the high-speed processor is also becoming a big problem. A dual-MAC processor with two ALU units may require 4 data words per clock cycle, or more than 4 gigabytes of data words per second.
(4) The challenges faced by the development of DSP are also reflected in the rapid increase in CPU speed and the continuous decline in price, which makes DSP manufacturers face two choices: one is to accelerate the development of DSP, and the other is to withdraw from the competition. Each manufacturer must shift from diversified investment to single investment and establish products with DSPS as the main development, that is, to integrate all technologies and all products into DSP.
4. Development Trends and Prospects of DSP
DSP continues to meet people's increasingly high requirements on its development path, and is gradually moving towards personalization and low power consumption. The prospects for the development of DSP are very impressive.
(1) System-level integrated DSP is the trend
. Reducing the size of DSP chips has always been the development direction of DSP technology. Most of the current DSPs are based on RISC (reduced instruction set computing) structure, which has the advantages of small size, low power consumption and high performance. Various DSP manufacturers have adopted new processes to improve the DSP core and integrate several DSP cores, MPU cores, dedicated processing units, peripheral circuit units and storage units on one chip to become a DSP system-level integrated circuit.
(2) Programmable DSP is the leading product.
Programmable DSP provides manufacturers with great flexibility. Manufacturers can develop a series of products of various models on the same DSP platform to meet the needs of different users. At the same time, programmable DSP also provides a good way for users to upgrade easily.
(3) Fixed-point DSP is the mainstream
. Theoretically, although the dynamic range of floating-point DSP is larger than that of fixed-point DSP and it is more suitable for DSP applications, the cost of fixed-point DSP devices is lower, the memory requirement is lower, and the power consumption is lower. Therefore, fixed-point programmable DSP devices are still the mainstream products on the market. According to statistics, the vast majority of DSP devices currently sold are 16-bit fixed-point programmable DSP devices, and it is expected that the proportion will gradually increase in the future.
(4) Pursuing higher computing speeds
At present, the general DSP computing speed is 100MIPS, that is, it can calculate 100 million instructions per second. Due to the personalization and customization trend of electronic devices, DSP must pursue higher and faster computing speeds to keep up with the pace of electronic device updates. The improvement of DSP computing speed mainly relies on new processes to improve chip structure. Most current DSP devices use 0.5μm~0.35μm CMOS process. According to the development trend of CMOS, it is entirely possible to increase the computing speed of DSP by 100 times (reaching 1600GIPS).
V. Conclusion
The growing application of DSP in various fields has driven the development of DSP itself. Its application in the 3C (Communications, Computers, Consumer) field has occupied 90% of the current DSP market share, indicating that DSP still has huge potential in other fields. In the future development, it will appear in various fields with better performance.

mdreamj

Great iron!!